ログイン 新規登録
言語:

WEKO3

  • トップ
  • ランキング
To
lat lon distance
To

Field does not validate



インデックスリンク

インデックスツリー

メールアドレスを入力してください。

WEKO

One fine body…

WEKO

One fine body…

アイテム

  1. シンポジウム
  2. シンポジウムシリーズ
  3. ゲームプログラミングワークショップ(GPWS)
  4. 2014

A Study of Software Framework for Parallel Monte Carlo Tree Search

https://ipsj.ixsq.nii.ac.jp/records/106508
https://ipsj.ixsq.nii.ac.jp/records/106508
236ecd36-6680-4348-b629-4bb1357ee4f7
名前 / ファイル ライセンス アクション
IPSJ-GPWS2014018.pdf IPSJ-GPWS2014018.pdf (1.4 MB)
Copyright (c) 2014 by the Information Processing Society of Japan
オープンアクセス
Item type Symposium(1)
公開日 2014-10-31
タイトル
タイトル A Study of Software Framework for Parallel Monte Carlo Tree Search
タイトル
言語 en
タイトル A Study of Software Framework for Parallel Monte Carlo Tree Search
言語
言語 eng
資源タイプ
資源タイプ識別子 http://purl.org/coar/resource_type/c_5794
資源タイプ conference paper
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者所属(英)
en
Department of Computer Science, National Chiao Tung University, Taiwan
著者名 Ting-FuLiao

× Ting-FuLiao

Ting-FuLiao

Search repository
I-ChenWu

× I-ChenWu

I-ChenWu

Search repository
Guan-WunChen

× Guan-WunChen

Guan-WunChen

Search repository
Chung-ChinShih

× Chung-ChinShih

Chung-ChinShih

Search repository
Po-YaKang

× Po-YaKang

Po-YaKang

Search repository
Bing-TsungChiang

× Bing-TsungChiang

Bing-TsungChiang

Search repository
Ting-ChuHo

× Ting-ChuHo

Ting-ChuHo

Search repository
Ti-RongWu

× Ti-RongWu

Ti-RongWu

Search repository
著者名(英) Ting-Fu, Liao

× Ting-Fu, Liao

en Ting-Fu, Liao

Search repository
I-Chen, Wu

× I-Chen, Wu

en I-Chen, Wu

Search repository
Guan-Wun, Chen

× Guan-Wun, Chen

en Guan-Wun, Chen

Search repository
Chung-Chin, Shih

× Chung-Chin, Shih

en Chung-Chin, Shih

Search repository
Po-Ya, Kang

× Po-Ya, Kang

en Po-Ya, Kang

Search repository
Bing-Tsung, Chiang

× Bing-Tsung, Chiang

en Bing-Tsung, Chiang

Search repository
Ting-Chu, Ho

× Ting-Chu, Ho

en Ting-Chu, Ho

Search repository
Ti-Rong, Wu

× Ti-Rong, Wu

en Ti-Rong, Wu

Search repository
論文抄録
内容記述タイプ Other
内容記述 Monte-Carlo tree search (MCTS) has been successful on improving the strength of the game Go as well as many other game playing programs. For MCTS, one of the critical issues to further improve strength is parallelization. In order to deal with parallel MCTS generally, this paper designs a software framework for developing computer game programs with parallel MCTS. This framework hides the details of game-independent designs from computer game developers, so that developers can concentrate on improving heuristics related to game-specific knowledge. In this framework, we used lock-free tree parallelization inside a shared-memory system, and root parallelization over a distributed-memory system. For demonstration, we implemented a Go program named AMIGO, a Chinese dark chess program, and a puzzle program on top of this framework. The experimental results of AMIGO for 9x9 Go also show reasonable speedups and the strength improvement.
論文抄録(英)
内容記述タイプ Other
内容記述 Monte-Carlo tree search (MCTS) has been successful on improving the strength of the game Go as well as many other game playing programs. For MCTS, one of the critical issues to further improve strength is parallelization. In order to deal with parallel MCTS generally, this paper designs a software framework for developing computer game programs with parallel MCTS. This framework hides the details of game-independent designs from computer game developers, so that developers can concentrate on improving heuristics related to game-specific knowledge. In this framework, we used lock-free tree parallelization inside a shared-memory system, and root parallelization over a distributed-memory system. For demonstration, we implemented a Go program named AMIGO, a Chinese dark chess program, and a puzzle program on top of this framework. The experimental results of AMIGO for 9x9 Go also show reasonable speedups and the strength improvement.
書誌情報 ゲームプログラミングワークショップ2014論文集

巻 2014, p. 122-126, 発行日 2014-10-31
出版者
言語 ja
出版者 情報処理学会
戻る
0
views
See details
Views

Versions

Ver.1 2025-01-21 09:29:43.780103
Show All versions

Share

Mendeley Twitter Facebook Print Addthis

Cite as

エクスポート

OAI-PMH
  • OAI-PMH JPCOAR
  • OAI-PMH DublinCore
  • OAI-PMH DDI
Other Formats
  • JSON
  • BIBTEX

Confirm


Powered by WEKO3


Powered by WEKO3